CivicActions has created this dashboard to be a visual representation of the analysis completed by Ben Balter, Staff Technical Program Manager at GitHub

This code can be found in its entirety in the CivicActions Data Science GitHub Repo.

---
title: "Digital Services Technology Dashboard"
output: 
  flexdashboard::flex_dashboard:
    orientation: rows
    social: menu
    source_code: embed
    theme: bootstrap
---

```{r setup, include=FALSE}
library(flexdashboard)
library(scales)
library(ggplot2)
library(ggmosaic)
library(plotly)
library(dplyr)
library(htmltools)
library(rmarkdown)

URL <- "https://ben.balter.com/2021-analysis-of-federal-dotgov-domains/data.csv"

dat <- read.csv(URL)
```

[CivicActions](www.civicactions.com) has created this dashboard to be a visual 
representation of the 
[analysis completed](https://ben.balter.com/2021/01/11/analysis-of-federal-dotgov-domains-pre-biden-edition/) 
by [Ben Balter](https://ben.balter.com/), Staff Technical Program Manager at GitHub

This code can be found in its entirety in the 
[CivicActions Data Science GitHub Repo](https://github.com/CivicActions/datascience/tree/main/data%20viz/digital_services_viz).

Row 
-----------------------------------------------------------------------

### Unique, Federally Managed `.gov` Domains

```{r}
valueBox(comma(length(unique(dat$domain_name))), icon = "fa-desktop")
```

### Unique, Federally Managed `.gov` Domains that are Live

```{r}
valueBox(comma(length(unique(dat$domain_name[dat$domain_up == "true"]))),
         icon = "fa-thumbs-up")
```

### `.gov` Domains that are Only Redirect to Another Domain

```{r}
valueBox(comma(length(unique(dat$domain_name[dat$domain_redirect == "true"]))),
         icon = "fa-unlink")
```

Row 
-----------------------------------------------------------------------

### Reachable Without the `www` Prefix

```{r}
valueBox(comma(length(unique(dat$domain_name[dat$domain_root == "true"]))), 
         icon = "fa-window-restore")
```

### `.gov` Domains Supporting IPV6

```{r}
valueBox(comma(length(unique(dat$domain_name[dat$dns_ipv6 == "true"]))), 
         icon = "fa-at")
```

### `.gov` Sites Using Open Source

```{r}
valueBox(comma(541), icon = "fa-folder-open")
```

Row {.tabset}
-----------------------------------------------------------------------

### Analytics

#### Of those live, non-redirected domains, 492 (53.13%) had a detectable Analytics tool.

```{r}
viz_dat <- case_when(
  grepl("Google Analytics.*and|and.*Google Analytics", dat$wappalyzer_analytics) ~ "Google Analytics + Other(s)",
  grepl("Google Analytics", dat$wappalyzer_analytics) ~ "Google Analytics",
  dat$wappalyzer_analytics == "" ~ "No Analytics Detected",
  1 == 1 ~ "Other",
)

viz_dat <- data.frame(analytics = c("Google Analytics + Other(s)", 
                                    "Google Analytics", 
                                    "No Analytics Detected", "Other"),
                      value = c(length(viz_dat[viz_dat == "Google Analytics + Other(s)"]),
                                length(viz_dat[viz_dat == "Google Analytics"]),
                                length(viz_dat[viz_dat == "No Analytics Detected"]),
                                length(viz_dat[viz_dat == "Other"])))

plot <- ggplot(viz_dat[viz_dat$analytics != "No Analytics Detected", ], 
               aes(x = reorder(analytics, -value), y = value)) + 
  geom_bar(stat = "identity", fill = "#D83933", 
             aes(text = sprintf("%s
# of sites: %s", analytics, value))) + 
  geom_text(label = viz_dat$value[viz_dat$analytics != "No Analytics Detected"], 
            nudge_y = 10, size = 4) + 
      theme(panel.background = element_blank(), 
          panel.grid.major.y = element_line(color = "light gray"),
          panel.grid.major.x = element_blank(), axis.ticks.x = element_blank(),
          axis.ticks.y = element_line(color = "light gray"), 
          axis.text = element_text(color = "black",size = 12),
          axis.title = element_blank(), 
          axis.text.y = element_text(color = "black"))

ggplotly(plot, tooltip = "text")
```

### Content Delivery Network

#### Of those live, non-redirected domains, 165 (17.82%) had a detectable CDN.

```{r}
viz_dat <- dat[dat$wappalyzer_cdn != "", ] %>%
    group_by(wappalyzer_cdn) %>%
    summarise(count = length(wappalyzer_cdn))

viz_dat$wappalyzer_cdn <- gsub("and", "and\n", viz_dat$wappalyzer_cdn)

plot <- ggplot(viz_dat, aes(x = reorder(wappalyzer_cdn, -count), y = count)) + 
    geom_bar(stat = "identity", fill = "#D83933", 
             aes(text = sprintf("%s
# of sites: %s", wappalyzer_cdn, count))) +
    geom_text(label = viz_dat$count, nudge_y = 3, size = 4) +
    theme(panel.background = element_blank(), 
          panel.grid.major.y = element_line(color = "light gray"),
          panel.grid.major.x = element_blank(), axis.ticks.x = element_blank(),
          axis.ticks.y = element_line(color = "light gray"), 
          axis.text.x = element_text(color = "black", hjust = TRUE, size = 12),
          axis.title = element_blank(), 
          axis.text.y = element_text(color = "black"))

ggplotly(plot, tooltip = "text")
```

### Content Management System

#### Of those live, non-redirected domains, 305 (32.94%) had a detectable CMS.

```{r}
viz_dat <- dat[dat$wappalyzer_cms != "", ] %>%
    group_by(wappalyzer_cms) %>%
    summarise(count = length(wappalyzer_cms))

tmp <- data.frame(wappalyzer_cms = "Other",
                  count = sum(viz_dat$count[viz_dat$count < 5]))

viz_dat <- viz_dat[viz_dat$count >= 5, ]
viz_dat <- rbind(viz_dat, tmp)
viz_dat$wappalyzer_cms <- case_when(
  viz_dat$wappalyzer_cms == "Microsoft SharePoint" ~ "Microsoft\nSharePoint",
  viz_dat$wappalyzer_cms == "Adobe Experience Manager" ~ "Adobe\nExp. Manager",
  1 == 1 ~ viz_dat$wappalyzer_cms
)

plot <- ggplot(viz_dat, aes(x = reorder(wappalyzer_cms, -count), y = count)) + 
    geom_bar(stat = "identity", fill = "#D83933", 
             aes(text = sprintf("%s
# of sites: %s", wappalyzer_cms, count))) +
    geom_text(label = viz_dat$count, nudge_y = 7, size = 4) +
    theme(panel.background = element_blank(), 
          panel.grid.major.y = element_line(color = "light gray"),
          panel.grid.major.x = element_blank(), axis.ticks.x = element_blank(),
          axis.ticks.y = element_line(color = "light gray"), 
          axis.text.x = element_text(color = "black", hjust = TRUE, size = 12),
          axis.title = element_blank(), 
          axis.text.y = element_text(color = "black"))

ggplotly(plot, tooltip = "text")
```

### Font Scripts

#### Of those live, non-redirected domains, 259 (27.09%) had a detectable Font Scripts.
#### *Note: Some sites use multiple Font Scripts.*

```{r}
viz_dat <- dat[dat$wappalyzer_font_scripts != "", ] %>%
    group_by(wappalyzer_font_scripts) %>%
    summarise(count = length(wappalyzer_font_scripts))

type <- data.frame(type = unique(unlist(strsplit(viz_dat$wappalyzer_font_scripts, " and "))))
type$search <- gsub(" \\(Twemoji\\)", "", type$type)

sum_dat <- data.frame()
for(font in type$search){
  tmp <- data.frame(type = type$type[type$search == font],
                    count = sum(viz_dat$count[grepl(font, viz_dat$wappalyzer_font_scripts)]))
  
  sum_dat <- rbind(sum_dat, tmp)
}

plot <- ggplot(sum_dat, aes(x = reorder(type, -count), y = count)) + 
    geom_bar(stat = "identity", fill = "#D83933", 
             aes(text = sprintf("%s
# of sites: %s", type, count))) +
    geom_text(label = sum_dat$count, nudge_y = 7, size = 4) +
    theme(panel.background = element_blank(), 
          panel.grid.major.y = element_line(color = "light gray"),
          panel.grid.major.x = element_blank(), axis.ticks.x = element_blank(),
          axis.ticks.y = element_line(color = "light gray"), 
          axis.text.x = element_text(color = "black", hjust = TRUE, size = 12),
          axis.title = element_blank(), 
          axis.text.y = element_text(color = "black"))

ggplotly(plot, tooltip = "text")
```

### JavaScript Frameworks

#### Of those live, non-redirected domains, 101 (10.91%) had a detectable JavaScript Framework.
#### *Note: Some sites use multiple JavaScript Frameworks.*
#### *Other Note: Other groups all JS Frameworks that were on less than 5 sites.*

```{r}
viz_dat <- dat[dat$wappalyzer_javascript_frameworks != "", ] %>%
    group_by(wappalyzer_javascript_frameworks) %>%
    summarise(count = length(wappalyzer_javascript_frameworks))

type <- unique(unlist(strsplit(viz_dat$wappalyzer_javascript_frameworks, " and ")))

sum_dat <- data.frame()
for(JSF in type){
  tmp <- data.frame(type = JSF,
                    count = sum(viz_dat$count[grepl(JSF, viz_dat$wappalyzer_javascript_frameworks)]))
  
  sum_dat <- rbind(sum_dat, tmp)
}

tmp <- data.frame(type = "Other",
                  count = sum(sum_dat$count[sum_dat$count < 5]))

sum_dat <- sum_dat[sum_dat$count >= 5, ]
sum_dat <- rbind(sum_dat, tmp)

plot <- ggplot(sum_dat, aes(x = reorder(type, -count), y = count)) + 
    geom_bar(stat = "identity", fill = "#D83933", 
             aes(text = sprintf("%s
# of sites: %s", type, count))) +
    geom_text(label = sum_dat$count, nudge_y = 1, size = 4) +
    theme(panel.background = element_blank(), 
          panel.grid.major.y = element_line(color = "light gray"),
          panel.grid.major.x = element_blank(), axis.ticks.x = element_blank(),
          axis.ticks.y = element_line(color = "light gray"), 
          axis.text.x = element_text(color = "black", hjust = TRUE, size = 12),
          axis.title = element_blank(), 
          axis.text.y = element_text(color = "black"))

ggplotly(plot, tooltip = "text")
```

### Platform as a Service

#### Of those live, non-redirected domains, 168 (18.14%) had a detectable PAAS.

```{r}
viz_dat <- dat[dat$wappalyzer_paas != "", ] %>%
    group_by(wappalyzer_paas) %>%
    summarise(count = length(wappalyzer_paas))
viz_dat$wappalyzer_paas <- gsub(" and ", "\nand ", viz_dat$wappalyzer_paas)

viz_dat$wappalyzer_paas <- gsub("Amazon Web Services", "AWS", 
                                viz_dat$wappalyzer_paas)

plot <- ggplot(viz_dat, aes(x = reorder(wappalyzer_paas, -count), y = count)) + 
    geom_bar(stat = "identity", fill = "#D83933", 
             aes(text = sprintf("%s
# of sites: %s", wappalyzer_paas, count))) +
    geom_text(label = viz_dat$count, nudge_y = 5, size = 4) +
    theme(panel.background = element_blank(), 
          panel.grid.major.y = element_line(color = "light gray"),
          panel.grid.major.x = element_blank(), axis.ticks.x = element_blank(),
          axis.ticks.y = element_line(color = "light gray"), 
          axis.text.x = element_text(color = "black", hjust = TRUE, size = 12),
          axis.title = element_blank(), 
          axis.text.y = element_text(color = "black"))

ggplotly(plot, tooltip = "text")
```

### Programming Languages

#### Of those live, non-redirected domains, 295 (31.86%) had a detectable Programming Language

```{r}
viz_dat <- dat[dat$wappalyzer_programming_languages != "", ] %>%
    group_by(wappalyzer_programming_languages) %>%
    summarise(count = length(wappalyzer_programming_languages))

viz_dat$wappalyzer_programming_languages <- gsub(" and ", "\nand ", 
                                                 viz_dat$wappalyzer_programming_languages)

plot <- ggplot(viz_dat, aes(x = reorder(wappalyzer_programming_languages, -count), y = count)) + 
    geom_bar(stat = "identity", fill = "#D83933", 
             aes(text = sprintf("%s
# of sites: %s", wappalyzer_programming_languages, count))) +
    geom_text(label = viz_dat$count, nudge_y = 5, size = 4) +
    theme(panel.background = element_blank(), 
          panel.grid.major.y = element_line(color = "light gray"),
          panel.grid.major.x = element_blank(), axis.ticks.x = element_blank(),
          axis.ticks.y = element_line(color = "light gray"), 
          axis.text.x = element_text(color = "black", hjust = TRUE, size = 12),
          axis.title = element_blank(), 
          axis.text.y = element_text(color = "black"))

ggplotly(plot, tooltip = "text")
```

### SSL/TLS certificate authorities

#### Of those live, non-redirected domains, 260 (28.08%) had a detectable SSL/TLS Certificate Authority.

```{r}
viz_dat <- dat[dat$wappalyzer_ssl.tls_certificate_authorities != "", ] %>%
    group_by(wappalyzer_ssl.tls_certificate_authorities) %>%
    summarise(count = length(wappalyzer_ssl.tls_certificate_authorities))

plot <- ggplot(viz_dat, aes(x = reorder(wappalyzer_ssl.tls_certificate_authorities, -count), y = count)) + 
    geom_bar(stat = "identity", fill = "#D83933", 
             aes(text = sprintf("%s
# of sites: %s", 
                                wappalyzer_ssl.tls_certificate_authorities, 
                                count))) +
    geom_text(label = viz_dat$count, nudge_y = 5, size = 4) +
    theme(panel.background = element_blank(), 
          panel.grid.major.y = element_line(color = "light gray"),
          panel.grid.major.x = element_blank(), axis.ticks.x = element_blank(),
          axis.ticks.y = element_line(color = "light gray"), 
          axis.text.x = element_text(color = "black", hjust = TRUE, size = 12),
          axis.title = element_blank(), 
          axis.text.y = element_text(color = "black"))

ggplotly(plot, tooltip = "text")
```

### Security

```{r}
viz_dat <- data.frame(Security = c("HTTPS Support", "HTST Support", 
                                "DNSSEC Record"),
                      value = c(dim(dat[dat$domain_https == "true", ])[1],
                                dim(dat[dat$domain_hsts == "true", ])[1],
                                dim(dat[dat$dns_dnssec == "true", ])[1]))

plot <- ggplot(viz_dat, aes(x = reorder(Security, -value), y = value)) + 
  geom_bar(stat = "identity", fill = "#D83933", 
             aes(text = sprintf("%s
# of sites: %s", Security, value))) + 
  geom_text(label = viz_dat$value, nudge_y = 35, size = 4) + 
      theme(panel.background = element_blank(), 
          panel.grid.major.y = element_line(color = "light gray"),
          panel.grid.major.x = element_blank(), axis.ticks.x = element_blank(),
          axis.ticks.y = element_line(color = "light gray"), 
          axis.text = element_text(color = "black",size = 12),
          axis.title = element_blank(), 
          axis.text.y = element_text(color = "black"))

ggplotly(plot, tooltip = "text")

```

### UI Frameworks

#### Of those live, non-redirected domains, 225 (24.30%) had a detectable UI Framework.
#### *Note: Some sites use multiple UI Frameworks.*

```{r}
viz_dat <- dat[dat$wappalyzer_ui_frameworks != "", ] %>%
    group_by(wappalyzer_ui_frameworks) %>%
    summarise(count = length(wappalyzer_ui_frameworks))

type <- unique(unlist(strsplit(viz_dat$wappalyzer_ui_frameworks, " and ")))

sum_dat <- data.frame()
for(JSF in type){
  tmp <- data.frame(type = JSF,
                    count = sum(viz_dat$count[grepl(JSF, viz_dat$wappalyzer_ui_frameworks)]))
  
  sum_dat <- rbind(sum_dat, tmp)
}

plot <- ggplot(sum_dat, aes(x = reorder(type, -count), y = count)) + 
    geom_bar(stat = "identity", fill = "#D83933", 
             aes(text = sprintf("%s
# of sites: %s", 
                                type, 
                                count))) +
    geom_text(label = sum_dat$count, nudge_y = 5, size = 4) +
    theme(panel.background = element_blank(), 
          panel.grid.major.y = element_line(color = "light gray"),
          panel.grid.major.x = element_blank(), axis.ticks.x = element_blank(),
          axis.ticks.y = element_line(color = "light gray"), 
          axis.text.x = element_text(color = "black", hjust = TRUE, size = 12),
          axis.title = element_blank(), 
          axis.text.y = element_text(color = "black"))

ggplotly(plot, tooltip = "text")
```


### Web Server

#### Of those live, non-redirected domains, 393 (42.44%) had a detectable Web Server.
#### *Note: Some sites use multiple Web Servers.*

```{r}
viz_dat <- dat[dat$wappalyzer_web_servers != "", ] %>%
    group_by(wappalyzer_web_servers) %>%
    summarise(count = length(wappalyzer_web_servers))

type <- unique(unlist(strsplit(viz_dat$wappalyzer_web_servers, " and ")))

sum_dat <- data.frame()
for(JSF in type){
  tmp <- data.frame(type = JSF,
                    count = sum(viz_dat$count[grepl(JSF, viz_dat$wappalyzer_web_servers)]))
  
  sum_dat <- rbind(sum_dat, tmp)
}

sum_dat$type <- gsub(" ", "\n", sum_dat$type)

plot <- ggplot(sum_dat, aes(x = reorder(type, -count), y = count)) + 
    geom_bar(stat = "identity", fill = "#D83933", 
             aes(text = sprintf("%s
# of sites: %s", 
                                type, 
                                count))) +
    geom_text(label = sum_dat$count, nudge_y = 5, size = 4) +
    theme(panel.background = element_blank(), 
          panel.grid.major.y = element_line(color = "light gray"),
          panel.grid.major.x = element_blank(), axis.ticks.x = element_blank(),
          axis.ticks.y = element_line(color = "light gray"), 
          axis.text.x = element_text(color = "black", hjust = TRUE, size = 12),
          axis.title = element_blank(), 
          axis.text.y = element_text(color = "black"))

ggplotly(plot, tooltip = "text")
```